Have a look at the example I have provided. There I use Connector/J V3.0.11-stable - which is the latest. If you use useUnicode=true and characterEncoding=UTF8 getString, does not return the correct string. At least on a jdk1.4.1_02 on a Solaris host. Any other oberservations?...
I haven't done these sort of things with MySQL so I cannot make specific comments. But I've seen weird things with other databases, where data was stored in the wrong encoding after having been transferred between systems, and you wouldn't notice when using the database's native tools.
So (and assuming ResultSet.getString() is expected to handle encoding correctly, which I only assume), in your case it could be either the driver, the database configuration or the actual data that causes the problem.
...OOH, with getByte and new String(byte, characterEncoding) it does. So it might be a problem with the JDBC Driver. But it's certainly more difficult to "fix" it in the Driver than in the cocoon source..
Sure, but adding JDBC driver workarounds to the Cocoon CVS for specific drivers does not seem right. What we could do is to add a configurable encoding that SQLTransformer would use to create Strings from Objects (or Byte []), but it still sounds like a hack.
-Bertrand
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
